Enhanced Customer Lifetime Value Prediction at Online Shopping

Customer lifetime value (CLV) is a critical component for businesses looking to maintain their clientele. Using a single data mining technique has its limitations when it comes to precisely calculating CLV in non-contractual situations. In order to overcome the issue, the research uses the customer lifecycle value theory to examine customer value evaluation and segmentation. The study employs Buy till you die (BTYD) models to enhance CLV predictions in non-contractual settings. These methods improve customer value segmentation and outperform traditional approaches. Through feature engineering a customer segmentation model is formed. The empirical study creates a model for customer segmentation in non-contractual contexts and segments customer value using machine learning algorithms and customer relationship management (CRM) analytical models. The usefulness and applicability of the suggested segmentation and value computation techniques are demonstrated through empirical validation, which is carried out using authentic transactional data from amazon. The customer segments were able to be explained to the recommend personalized marketing campaigns, loyalty programs, enhanced support, and other activities targeted at increasing customer satisfaction and long-term value. It is also suggested that businesses should prioritize high-value customers and find a way to uplift the moderate spenders through focused strategies, including cross-selling or bundling. This paper suggests an enhanced model tested and experimented with the performance on classifying and predicting real-time data for predicting the customer lifetime value when the customers are with infrequent buying behavior.